Using job posting and resume data to better understand the U.S. labor market
The U.S. labor market literature relies, for the most part, on government survey data, which is invaluable for identifying economywide trends and differentials but lacks the detail needed about tasks and occupations to detect changes in how work is organized. This project will use two datasets assembled by Burning Glass Technologies containing millions of job postings and millions of resumes. The data offer the scope and scale to make fine distinctions and explore the array of skills embodied in workers and the tasks called out by employers.
This project has three aims. First, it will examine variation within occupational categories and connections across categories. Given the growth of within-occupational inequality, understanding within-occupation heterogeneity in skills and other attributes is extremely important. Second, it will use resume data to examine worker mobility. This is a novel approach and is likely to offer important insights about labor market mobility. Third, it will investigate job quality, including nonwage benefits and employment relations, and the incidence of nonstandard work (temporary, on call, and contract based). Nonstandard work is not regularly measured in government surveys, so exploring whether job posting data can offer insight into the prevalence of nonstandard work arrangements and whether such arrangements are associated with job mobility and access to nonwage benefits, as well as differences in wage levels between similar occupations with different employment relations, could lay the groundwork for future research.